Self-Supervised Learning of Monocular Visual Odometry and Depth with Uncertainty-Aware Scale Consistency
Changhao Wang,Guanwen Zhang,Wei Zhou,Changhao Wang,Guanwen Zhang,Wei Zhou
The inherent scale ambiguity issue greatly limits the performance of monocular visual odometry. In recent years, a variety of methods have been proposed for self-supervised learning of ego-motion and depth estimation, incorporating specifically designed scale-consistency constraints that utilize estimated depth as a reference. However, these existing methods neglect the influence of the depth unce...
